diff --git a/.cursor/skills/test-run/SKILL.md b/.cursor/skills/test-run/SKILL.md
index 0bd22d6..adec4c6 100644
--- a/.cursor/skills/test-run/SKILL.md
+++ b/.cursor/skills/test-run/SKILL.md
@@ -21,8 +21,8 @@ Run the project's test suite and report results. This skill is invoked by the au
 
 Check in order — first match wins:
 
-1. `scripts/run-tests.sh` exists → use it
-2. `docker-compose.test.yml` or equivalent test environment exists → spin it up first, then detect runner below
+1. `scripts/run-tests.sh` exists → use it (the script already encodes the correct execution strategy)
+2. `docker-compose.test.yml` exists → run the Docker Suitability Check (see below). Docker is preferred; use it unless hardware constraints prevent it.
 3. Auto-detect from project files:
    - `pytest.ini`, `pyproject.toml` with `[tool.pytest]`, or `conftest.py` → `pytest`
    - `*.csproj` or `*.sln` → `dotnet test`
@@ -32,6 +32,37 @@ Check in order — first match wins:
 
 If no runner detected → report failure and ask user to specify.
 
+#### Docker Suitability Check
+
+Docker is the preferred test environment. Before using it, verify no constraints prevent easy Docker execution:
+
+1. Check `_docs/02_document/tests/environment.md` for a "Test Execution" decision (if the test-spec skill already assessed this, follow that decision)
+2. If no prior decision exists, check for disqualifying factors:
+   - Hardware bindings: GPU, MPS, CUDA, TPU, FPGA, sensors, cameras, serial devices, host-level drivers
+   - Host dependencies: licensed software, OS-specific services, kernel modules, proprietary SDKs
+   - Data/volume constraints: large files (> 100MB) impractical to copy into a container
+   - Network/environment: host networking, VPN, specific DNS/firewall rules
+   - Performance: Docker overhead would invalidate benchmarks or latency measurements
+3. If any disqualifying factor found → fall back to local test runner. Present to user using Choose format:
+
+```
+══════════════════════════════════════
+ DECISION REQUIRED: Docker is preferred but factors
+ preventing easy Docker execution detected
+══════════════════════════════════════
+ Factors detected:
+ - [list factors]
+══════════════════════════════════════
+ A) Run tests locally (recommended)
+ B) Run tests in Docker anyway
+══════════════════════════════════════
+ Recommendation: A — detected constraints prevent
+ easy Docker execution
+══════════════════════════════════════
+```
+
+4. If no disqualifying factors → use Docker (preferred default)
+
 ### 2. Run Tests
 
 1. Execute the detected test runner
diff --git a/.cursor/skills/test-spec/SKILL.md b/.cursor/skills/test-spec/SKILL.md
index 73f3890..cdd6a7a 100644
--- a/.cursor/skills/test-spec/SKILL.md
+++ b/.cursor/skills/test-spec/SKILL.md
@@ -209,7 +209,7 @@ Based on all acquired data, acceptance_criteria, and restrictions, form detailed
 - [ ] Expected results use comparison methods from `.cursor/skills/test-spec/templates/expected-results.md`
 - [ ] Positive and negative scenarios are balanced
 - [ ] Consumer app has no direct access to system internals
-- [ ] Docker environment is self-contained (`docker compose up` sufficient)
+- [ ] Test environment matches project constraints (see Docker Suitability Assessment below)
 - [ ] External dependencies have mock/stub services defined
 - [ ] Traceability matrix has no uncovered AC or restrictions
 
@@ -337,11 +337,51 @@ When coverage ≥ 70% and all remaining tests have validated data AND quantifiab
 
 ---
 
+### Docker Suitability Assessment (BLOCKING — runs before Phase 4)
+
+Docker is the **preferred** test execution environment (reproducibility, isolation, CI parity). Before generating scripts, check whether the project has any constraints that prevent easy Docker usage.
+
+**Disqualifying factors** (any one is sufficient to fall back to local):
+- Hardware bindings: GPU, MPS, TPU, FPGA, accelerators, sensors, cameras, serial devices, host-level drivers (CUDA, Metal, OpenCL, etc.)
+- Host dependencies: licensed software, OS-specific services, kernel modules, proprietary SDKs not installable in a container
+- Data/volume constraints: large files (> 100MB) that would be impractical to copy into a container, databases that must run on the host
+- Network/environment: tests that require host networking, VPN access, or specific DNS/firewall rules
+- Performance: Docker overhead would invalidate benchmarks or latency-sensitive measurements
+
+**Assessment steps**:
+1. Scan project source, config files, and dependencies for indicators of the factors above
+2. Check `TESTS_OUTPUT_DIR/environment.md` for environment requirements
+3. Check `_docs/00_problem/restrictions.md` and `_docs/01_solution/solution.md` for constraints
+
+**Decision**:
+- If ANY disqualifying factor is found → recommend **local test execution** as fallback. Present to user using Choose format:
+
+```
+══════════════════════════════════════
+ DECISION REQUIRED: Test execution environment
+══════════════════════════════════════
+ Docker is preferred, but factors preventing easy
+ Docker execution detected:
+ - [list factors found]
+══════════════════════════════════════
+ A) Local execution (recommended)
+ B) Docker execution (constraints may cause issues)
+══════════════════════════════════════
+ Recommendation: A — detected constraints prevent
+ easy Docker execution
+══════════════════════════════════════
+```
+
+- If NO disqualifying factors → use Docker (preferred default)
+- Record the decision in `TESTS_OUTPUT_DIR/environment.md` under a "Test Execution" section
+
+---
+
 ### Phase 4: Test Runner Script Generation
 
 **Role**: DevOps engineer
 **Goal**: Generate executable shell scripts that run the specified tests, so the autopilot and CI can invoke them consistently.
-**Constraints**: Scripts must be idempotent, portable across dev/CI, and exit with non-zero on failure.
+**Constraints**: Scripts must be idempotent, portable across dev/CI, and exit with non-zero on failure. Respect the Docker Suitability Assessment decision above.
 
 #### Step 1 — Detect test infrastructure
 
@@ -350,7 +390,9 @@ When coverage ≥ 70% and all remaining tests have validated data AND quantifiab
    - .NET: `dotnet test` (*.csproj, *.sln)
    - Rust: `cargo test` (Cargo.toml)
    - Node: `npm test` or `vitest` / `jest` (package.json)
-2. Identify docker-compose files for integration/blackbox tests (`docker-compose.test.yml`, `e2e/docker-compose*.yml`)
+2. Check Docker Suitability Assessment result:
+   - If **local execution** was chosen → do NOT generate docker-compose test files; scripts run directly on host
+   - If **Docker execution** was chosen → identify/generate docker-compose files for integration/blackbox tests
 3. Identify performance/load testing tools from dependencies (k6, locust, artillery, wrk, or built-in benchmarks)
 4. Read `TESTS_OUTPUT_DIR/environment.md` for infrastructure requirements
 
@@ -360,10 +402,11 @@ Create `scripts/run-tests.sh` at the project root using `.cursor/skills/test-spe
 
 1. Set `set -euo pipefail` and trap cleanup on EXIT
 2. Optionally accept a `--unit-only` flag to skip blackbox tests
-3. Run unit tests using the detected test runner
-4. If blackbox tests exist: spin up docker-compose environment, wait for health checks, run blackbox test suite, tear down
-5. Print a summary of passed/failed/skipped tests
-6. Exit 0 on all pass, exit 1 on any failure
+3. Run unit/blackbox tests using the detected test runner:
+   - **Local mode**: activate virtualenv (if present), run test runner directly on host
+   - **Docker mode**: spin up docker-compose environment, wait for health checks, run test suite, tear down
+4. Print a summary of passed/failed/skipped tests
+5. Exit 0 on all pass, exit 1 on any failure
 
 #### Step 3 — Generate `scripts/run-performance-tests.sh`
 
@@ -371,7 +414,7 @@ Create `scripts/run-performance-tests.sh` at the project root. The script must:
 
 1. Set `set -euo pipefail` and trap cleanup on EXIT
 2. Read thresholds from `_docs/02_document/tests/performance-tests.md` (or accept as CLI args)
-3. Spin up the system under test (docker-compose or local)
+3. Start the system under test (local or docker-compose, matching the Docker Suitability Assessment decision)
 4. Run load/performance scenarios using the detected tool
 5. Compare results against threshold values from the test spec
 6. Print a pass/fail summary per scenario
diff --git a/_docs/_autopilot_state.md b/_docs/_autopilot_state.md
index 01e7b1d..5bd50d2 100644
--- a/_docs/_autopilot_state.md
+++ b/_docs/_autopilot_state.md
@@ -4,6 +4,6 @@
 flow: existing-code
 step: 6
 name: Run Tests
-status: failed
+status: in_progress
 sub_step: 0
-retry_count: 0
+retry_count: 1